The flatness, bow, warp and other qualities of semiconductor wafers must be within precise standards in order for the wafers to be usable. Individual wafers are measured to determine whether and the extent of conformance to such standards. Errors are, however, introduced into the measurements by the mechanical errors of the fixtures holding the wafers, thereby limiting the accuracy of the wafer measurement. One solution in high accuracy scenarios is the use of wafer test fixtures that are so precisely fabricated as to exhibit substantially no error themselves. This solution is disadvantageous, however, due to the limited processing throughput that such devices often entail and due to the expenses that are associated with providing and maintaining such test fixtures. Another solution is to electronically process the measurement data to compensate the measurement data for the fixture induced errors to the extent that they are repeatable. In commonly assigned copending U.S. utility patent application Ser. No. 802,049, incorporated herein by reference, now U.S. Pat. No. 4,750,141 method and apparatus are disclosed for separating fixture errors from measured data, and for compensating the measured data for the error introduced by the fixture error. In a bow and warp station embodiment thereof, an X, .theta., and Z movable fixture is calibrated to determine its X and .theta. fixture induced errors that repeat measurement cycle to measurement cycle. Once the fixture is calibrated, the measurement data is therewith compensated providing an accuracy not limited by the repeatable X, .theta. errors associated with the mechanical nature of the fixture.